Join us on June 5th for a Tech Talk with Bill Doerrfeld and Kenn Hussey as we discuss the future of open source. Register now
Back to blog

Getting Started with Telepresence on Windows

Kay James
January 31, 2017 | 8 min read
Telepresence on Windows

Telepresence is a powerful open source tool that allows you to code and test a service locally against a remote Kubernetes cluster. This allows you to quickly build and debug a local service that is dependent on a large number of remote services and resource-hungry remote middleware or databases. Telepresence is now available in developer preview for Windows and in this guide, you will learn how to install Telepresence and create an “intercept” to route traffic from your cluster to your local environment on Windows machines.

Before you begin

This article assumes that you have your own remote cluster with an application deployed that uses the microservices architecture and you want to intercept (or re-route) traffic to a service in that cluster. If you do not have your own cluster but would like to try Telepresence you can follow this quickstart with a free demo cluster from Ambassador Labs.


  • A Kubernetes cluster with a service you want to intercept
  • kubectl (OpenShift users can use oc)
  • Windows Powershell running as an administrator

Install Telepresence

Before you can start taking advantage of Telepresence, you need to install the application. There are a couple methods which you can choose to download the install file. At the moment, Telepresence for Windows is installed via a powershell script; future versions of Telepresence for Windows will be installed as a Windows exe. Newer Windows 10 machines (build 17063 or later) come with the “curl” command. If your computer has curl you can download Telepresence using this command in your Powershell session:

curl -fL -o

In the case where you do not have curl installed you can download the install file using Invoke-WebRequest:

Invoke-WebRequest -o

After the file is downloaded you’ll need to unzip it and clean it up by running the following commands:

Expand-Archive -Path
Remove-Item ''
cd telepresence

Run the install-telepresence.ps1 to install telepresence's dependencies. By default Telepresence will be installed at C:/telepresence, but you may specify a custom path $path with -Path $path.

Set-ExecutionPolicy Bypass -Scope Process

A prompt will come up asking if you want to change the execution policy. Enter (Y) here and the install process will continue.

change the execution policy
change the execution policy

Finally remove the unzipped file with:

cd ..
Remove-Item telepresence

Congratulations, you have successfully installed Telepresence on your Windows machine. Before moving on, make sure to close your Powershell and open a new one.

Connect to your remote cluster

Now that you have installed Telepresence and opened a new Powershell session, you can begin to connect Telepresence to your cluster and start intercepting your services. First make sure that you are in the correct context with the cluster and namespace of the services you’d like to intercept.

1. Use kubectl get services and make sure that you’re in the right namespace for the service you want to intercept. In my case, the “my-context” defaults to the namespace “foo” where I have my service called “quote."


2. Next run the

telepresence connect

run the telepresence connect command
run the telepresence connect command

3. The

telepresence list
command will display the services that are available to intercept in the current namespace. You can add
-n <namespace>
to the end of the command to show services in another namespace without changing your context.

add -n <namespace> to the end of the command
add -n <namespace> to the end of the command

In my example, the quote service displays a random entry from a list of quotes. With a “curl” request to the quote service, we can access the remote service and see one of the entries from the list of quotes. The Telepresence DNS resolver dynamically resolves names using namespaces; access your remote service locally following the format of this command:

curl <service_name>.<namespace>

Replace <service_name> with the name of the service you want to intercept as listed in

telepresence list
and <namespace> with the namespace that service is in.


Run your service locally

Before you actually intercept your remote service, you should run a version of the service locally. Make a change to your local service in order to compare with the service in the cluster.

In my example, my local instance of the quote service has a new set of quotes. My service is deployed via Docker containers so I will create a new build and run the service using docker run.

After you edit the code of your own local service, create a new build and run your service.


Take note of the port your local instance of the service is running on. It will be important in the next step. In my example, the quote service is running on localhost:8080 and returns one of the new quotes updated in the new build.

quote service is running on localhost:8080
quote service is running on localhost:8080

Create an intercept

Now that you have the service running locally, you can create an intercept with Telepresence to intercept traffic going to your service in the cluster and reroute it to your local instance. There are two methods of intercepting traffic you can do: intercept all traffic to the service or intercept certain requests to your service using a preview URL. In this example, we will use preview URLs to create a personal intercept that is shareable via a preview URL.

Telepresence login

In order to use preview URLs you need to create an account here. Run

telepresence login
 to open your browser and login using your preferred credentials (Github, Gitlab, or Google).

Follow the login process and return to your terminal:

Telepresence intercept

After logging in, you can create a personal intercept with the following command:

telepresence intercept <service_name> --port <local_port>

Replace <service_name> with the name of your service (as listed in telepresence list) and <local_port> with the port you indicated in the previous step.

Telepresence needs to know how your cluster is configured to properly set up the intercept, so it will ask you four questions about how your cluster ingress works.

  1. Your ingress layer 3 address: this will usually be the internal address of your ingress controller using the format <service_name>.namespace.
  2. Ingress layer 4 address: the port which your ingress controller is listening (usually 443 for TLS and 80 for non-TLS)
  3. Ingress TLS encryption: whether your ingress controller expects TLS communication on the specified port or not
  4. Layer 5 hostname: If your ingress routes traffic via a host name enter it here

After answering the questions, telepresence will create the intercept and return a preview URL.

With the preview URL, traffic that was going to your remote service is now rerouted to the service running locally and you can test your changes as if they were already deployed in the cluster. I can access my quote service using the same prefix I normally would in the cluster.

service running locally
service running locally

Notice how the remote service in your cluster has been unaffected. In my example, it still returns one of the old quotes.

remote service in a cluster
remote service in a cluster

Telepresence’s preview URLs allow you to test and debug changes to your service without affecting the cluster or anyone else who may be working with the cluster. The preview URL can also be shared and is accessible to others, so collaboration becomes incredibly easy and simple.

Congratulations, you’ve successfully installed Telepresence and created an “intercept” routing traffic from your cluster to your local environment!


If you encounter errors while trying to create or access your preview URL, you may have configured your intercept incorrectly. Use

telepresence leave <service_name>
to remove the intercept and then create a new intercept again. Make sure to answer the four questions with the correct configurations for your cluster. Telepresence tries to determine these settings for you, but the given defaults may not always be appropriate for your cluster.

Find more troubleshooting tips here.